Conservation in decay accelerating factor (DAF) structure among primates

Developmental and Comparative Immunology 24 (2000) 815±827 www.elsevier.com/locate/devcompimm

Conservation in decay accelerating factor (DAF) structure among primates Lisa Kuttner-Kondo a, V. Bala Subramanian b, John P. Atkinson b, Jianliang Yu a, M. Edward Medof a,* a

Department of Pathology, Case Western Reserve University, School of Medicine, Institute of Pathology, 2085 Adelbert Road, Rm. 301, Cleveland, OH 44106, USA b Department of Internal Medicine, Washington University School of Medicine, 4950 Childrens Place, St. Louis, MO 63110, USA Received 10 August 1999; received in revised form 21 February 2000; accepted 2 March 2000

Abstract The decay accelerating factor (DAF, CD55) protects self cells from activation of autologous complement on their surfaces. It functions to disable the C3 convertases, the central ampli®cation enzymes of the cascade. Its active site(s) are contained within four 060 amino acid long units, termed complement control protein repeats (CCPs), which are suspended above the cell surface on a 68 amino acid long serine/threonine (S/T)-rich cushion that derives from three exons. We previously proposed a molecular model of human DAF's four CCPs in which certain amino acids were postulated to be recognition sites for the interaction between DAF and the C3 convertases. In the current study, we characterized DAF in ®ve non-human primates: the great apes, gorilla and common chimpanzee, and the Old World monkeys: hamadryas baboon, Rhesus macaque, and patas monkey. Amino acid homology to human DAF was approximately 98% for the two great apes and 83% for the three Old World monkeys. The above cited putative ligand interactive residues were found to be fully conserved in all of the non-human primates, although there were amino acid changes outside of these areas. In the chimpanzee, alternative splicing of the S/T region was found potentially to be the source of multiple protein isoforms in erythrocytes, whereas in the patas monkey, similar alternative splicing was observed but only one protein band was seen. Interestingly, a Rhesus macaque was found to exhibit a phenomenon paralleling the human Cromer Dr(a-) blood group, in which a 44-base pair deletion in CCP3 leads to a frameshift and early STOP codon. 7 2000 Elsevier Science Ltd. All rights reserved. Keywords: DAF; CD55; Complement; Primates; C3 convertases; Recognition of self; Xenotransplants; Cancer; Cell surface regulatory proteins

Abbreviations: DAF, decay accelerating factor; CCPs, complement control protein repeats; S/T, serine/threonine; GPI, glycosylphosphatidylinositol; FBS, fetal bovine serum; PBMCs, peripheral blood mononuclear cells; RT, reverse transcription; mAb, monoclonal antibody; Ehu, human erythrocyte; UTR, untranslated region. * Corresponding author. Tel.: +1-216-368-5434; fax: +1-216-368-0495. E-mail address: [email protected] (M.E. Medof). 0145-305X/00/$ - see front matter 7 2000 Elsevier Science Ltd. All rights reserved. PII: S 0 1 4 5 - 3 0 5 X ( 0 0 ) 0 0 0 2 6 - 4

816

L. Kuttner-Kondo et al. / Developmental and Comparative Immunology 24 (2000) 815±827

1. Introduction The decay accelerating factor (DAF, CD55) is a cell surface regulatory protein that functions to protect host cells from autologous complement attack. It is widely expressed, being present not only on blood and vascular endothelial cells that are in intimate contact with complement, but also on most extravascular cell types [1]. It downregulates complement activation on self cells by dissociating the components of the alternative and classical pathway C3 convertases, C3bBb and C4b2a [2]. It is composed of four 060 amino acid long repeating units, termed complement control protein repeats (CCPs), followed by a serine/threonine (S/T)-rich region [3±5] which in turn is linked to the membrane by a posttranslationally added glycosylphosphatidylinositol (GPI) anchor. Its dissociative capacity lies within its four CCPs, while the S/T region acts as an antiproteolytic spacer that appropriately positions these functional domains above host cell membranes [6,7]. DAF's structure is conserved among all mammals studied so far. It has been characterized in the orangutan [8], mouse [9], guinea pig [10], and rat [11] in addition to humans. In the CCP functional domains, human and orangutan DAF exhibit 93% identity at the amino acid level. Comparisons of human with mouse, guinea pig and rat DAF show 50±60% amino acid identity. In humans, orangutans, guinea pigs, and rats, there is a single copy of the DAF gene [8,10±12]. Despite this, in the guinea pig, alternative splicing of the three exons (equivalent to human A, B and C) encoding the S/T region and the exon(s) encoding the C-terminus gives rise to multiple DAF isoforms, including GPI-anchored and transmembrane variants each with variable length S/T regions, as well as a secreted DAF protein [10,13]. The rat makes GPI-anchored, transmembraneanchored and secreted DAF forms [11,14]. It is not yet clear whether more than one DAF isoform exists in humans [15] or orangutans [8]. The mouse has two DAF genes, one encoding a GPI-anchored protein, the other a transmembrane protein [9]. Recently, sequen-

cing [16] of intron 7, and portions of exon 7 (corresponding to S/T region A) of the human DAF gene [17] in seven non-human primates including representatives of the great apes, lesser apes, Old World monkeys, and New World monkeys, as well as in the guinea pig and mouse has shown a unique repeating unit. In all of these species, exon 7 derives from two or more of these units, indicating that the N-terminal portion of the S/T region evolved prior to diversi®cation. In the guinea pig, their translation underlies the production of the alternative isoforms of DAF protein [16]. In previous studies, we proposed a molecular model of DAF's four CCPs in which a positively charged surface area on CCPs 2 and 3 was postulated to be the primary recognition site for C3bBb and C4b2a. Several nearby accessible hydrophobic amino acids, V121, F123, L147 and F148, were postulated to aid in the interaction between DAF and the C3 convertases. Two surface depressions on CCPs 3 and 4 were proposed to provide additional ligand binding sites [18]. To further investigate the importance of these areas in DAF's function, the current study reports on the molecular cloning of DAF homologues from ®ve non-human primates: two great apes, Pan troglodytes (common chimpanzee) and Gorilla gorilla (gorilla) and three Old World monkeys, Papio hamadryas (hamadryas baboon), Macaca mulatta (Rhesus macaque) and Erythrocebus patas (patas monkey). We found that in all of these species, the above putative convertase interactive areas are entirely conserved. We additionally found that the chimpanzee and the patas monkey express mRNAs encoding multiple DAF isoforms stemming from alternative splicing of the S/T-rich region. 2. Methods and materials 2.1. Cell culture and puri®cation The baboon lymphoblastoid cell line 26CB1 and the Rhesus macaque normal kidney epithelial cell line NCTC clone 3526 were obtained

L. Kuttner-Kondo et al. / Developmental and Comparative Immunology 24 (2000) 815±827

from the ATCC. 26CB1 cells were grown in RPMI 1640/L-glutamine medium supplemented with fetal bovine serum [FBS (10%)], and penicillin (100 U/ml)/streptomycin (100 mg/ml). NCTC clone 3526 was grown in NCTC 109/L-glutamine medium (BioWhittaker, Walkersville, MD) containing horse serum (10%), and penicillin (100 U/ml)/streptomycin (100 mg/ml). Blood from male patas monkeys #I983 and #B289 and male patas monkey #R181 was obtained from Tulane Regional Primate Research Center, Tulane University, New Orleans, LA and from Bioqual, Inc., Rockville, MD, respectively. Blood from male Rhesus macaque #OPE340 was obtained from Yerkes Regional Primate Research Center, Emory University, Atlanta, GA. Peripheral blood mononuclear cells (PBMCs) were puri®ed by Ficoll±Paque density centrifugation. 2.2. RNA and genomic DNA isolation Total RNA was obtained from the Pan troglodytes LENA B cell line (chimpanzee), Gorilla gorilla ROK B cell line (gorilla), Papio hamadryas 26CB1 lymphoblastoid cell line (baboon), Macaca mulatta OPE340 PBMCs (Rhesus macaque), and Erythrocebus patas PBMCs (patas monkey). Additional total RNA was isolated from 26CB1, NCTC clone 3526, and patas monkey leukocytes using the Tripure Isolation method (Boehringer Mannheim, Indianapolis, IN) and the products redissolved in diethyl pyrocarbonate-treated water. Prior to RNA isolation, 26CB1 cells were resuspended in PBS. In the case of NCTC clone 3526, Tripure reagent was applied at 3 ml/100 cm2 plate. Genomic DNA was prepared using the Wizard Genomic DNA Puri®cation Kit (Promega, Madison, WI). 2.3. Reverse transcription (RT)-PCR RT was performed using a Superscript protocol (Life Technologies, Gaithersburg, MD). PCR was conducted initially using human and subsequently non-human DAF primers in conjunction with non-proofreading (Taq, Promega, Madison, WI; and Display Taq, PGC Scienti®cs,

817

Gaithersburg, MD) and proofreading (Ultma, Perkin-Elmer, Foster City, CA) DNA polymerases. For each primate, conditions were optimized but never exceeded 2 mM Mg2+, 0.2 mM of each dNTP, 5U polymerase and 50 pmol of each primer per 100 ml reaction. Initial denaturation was performed at 948C for 3 min, followed by 30 or 35 cycles of denaturation of 948C for 1 min, annealing at 558C for 1 min and 15 s, extension at 728C for 1 min and 15 s, and a ®nal extension at 728C for 7 min. In some cases a lower annealing temperature (Rhesus, 508C) or a greater number of cycles was used (patas, 40). PCR products were analyzed on 1% agarose gels. The primer pairs employed for each nonhuman primate are listed in Table 1. Those used for generation of the longest DAF sequences were D185 (B, C, M) and D1232R or D1451R. They encompassed nucleotides encoding residue eight of CCP1 through four nucleotides or 0200 nucleotides, respectively, downstream of the stop codon. Additional primer pairs were used to con®rm the ®rst seven amino acids of CCP1. All were derived from human DAF sequence with the exception of (i) D185B and D914RB which were derived from baboon DAF, (ii) D185C and D1191RC which were derived from chimpanzee DAF, and (iii) D185M, D77BM, and D914RBM which were derived from Rhesus DAF (numbers are relative to human DAF sequence). The sequences of the 5' DAF primers (5 ' to 3 ') were: 1. D1-ACTGCAACTCGCTCCGGCCGCTGGGCGTAG (transcription start); 2. D28-TAGCTGCGACTCGGCGGAGT (28 bp downstream of the transcription start site); 3. D77BM-CGCGCCATGACTGTCGCGCGGCCGAGCGTG (77 bp downstream of the transcription start site); 4. D185-GACTGTGGCCTTCCCCCAGAT (5 ' end of CCP1); 5. D185B-GACTGTGGCCCTCCCCCAGCT (5 ' end of CCP1); 6. D185C-GACTGTGGCCTTCCCCCAGAG (5 ' end of CCP1); 7. D185M-GACTGTGGCCCTCCCCCCGCT

818

L. Kuttner-Kondo et al. / Developmental and Comparative Immunology 24 (2000) 815±827

(5 ' end of CCP1); 8. D 2 0 6 - G T A C C T A A T G C C C A G C C A G C (within CCP1); 9. D368-CGTAGCTGCGAGGTGCCAACAA GGCTAAAT (5 ' end of CCP2); 10. D758-CCAGCACCACCACAAATTGA (within CCP4); and 11. D959-CCACCAACAGTTCAGAAACCTA CCACAGTA (within S/T-A). The sequences of the 3 ' DAF primers (5 ' to 3') were: 1. D338R-ATTGCAGAACTCTTCAATATCT GACCATTG (near the 3' end of CCP1); 2. D914R-TCTGCATTCAGGTGGTGGGCC (3 ' end of CCP4); 3. D914RB-TCTACATGCAGGTGGTGGGCC (3 ' end of CCP4); 4. D914RBM-GTTTGCTCTACATGCAGGTG GTGGGCC (3 ' end of CCP4); 5. D1191RC-GGTTACTAGCATCCCAAGCA (within the C-terminal hydrophobic region); 6. D1232R-GTGTGTATTTTCTTCTTAACTC TTCTT (4 nucleotides downstream of the stop codon); and 7. D1451R-TTTTAGGAAAGGAATCACTCTC AATTCTGC [within the 3 ' untranslated region (UTR)].

2.4. Subcloning and sequencing of the PCR products PCR products were puri®ed either by phenol extraction from low-melting point agarose fol-

lowed by NaOAc/Ethanol precipitation or by using the JETQUICK Gel Extraction Spin Kit (Genomed, Research Triangle Park, NC). Products were ligated into the pT7B (R) vector using the Novagen Blunt Cloning Kit and sequenced on an ABI Prism 377 DNA Sequencer using the ``BigDye'' Terminator Cycle Sequencing Kit and the Rhodamine Matrix program. For each species, at least two independent PCR analyses were carried out. 2.5. Western blot analyses Primate DAF proteins in erythrocyte lysates were electrophoresed on 7.5% or 10% SDSPAGE nonreducing gels. Separated proteins were transferred using a Transblot apparatus (Biorad, Hercules, CA) and revealed with rabbit polyclonal antibodies (Medof, ME, unpublished) or murine monoclonal antibody (mAb) IA10 to human erythrocyte (Ehu) DAF [19]. 3. Results 3.1. Primate DAF proteins Fig. 1(panel A) shows a comparative Western blot analysis of primate erythrocyte and Ehu DAF proteins as revealed by using rabbit polyclonal antibody raised against Ehu DAF protein. Similar to Ehu DAF, a major band of 60±70 kDa is seen for each of the great apes (chimpanzee and gorilla) and each of the Old World monkeys (baboon, Rhesus macaque and patas monkey). A

Table 1 Primer pairs used in PCR Non-primate cells

Primer pairs

ROK LENA 26CB1 OPE340 PBMCs Patas R181 leukocytes NCTC3526 Patas B289 PBMCs

D185-D1451R, D28-D338R, D1-D338R, D185-D1232R, D959-D1451R D185-D1451R, D28-D338R, D1-D338R, D758-D1191RC, D185C-D1232R, D185C-D914R, D758-D1451R D185-D1451R, D28-D338R, D185B-D1232R, D185B-D1451R D206-D914R, D368-D1451R, D28-D338R D28-D338R, D185-D1232R, D758-D1451R, D185B-D1232R, D185B-D1451R D28-D338R, D77BM-D914RBM, D206-D914RB, D758-D1451R, D185M-D1232R, D959-D1451R D28-D338R, D185-D914RB, D758-D1451R, D185B-D1451R, D185B-D1232R

L. Kuttner-Kondo et al. / Developmental and Comparative Immunology 24 (2000) 815±827

higher molecular weight band, similar to the human DAF dimer (termed DAF2), additionally is seen for chimpanzee and gorilla, while two lower molecular weight bands are seen for chimpanzee.

819

As shown in Fig. 1(panel B), mouse mAb IA10 [19], directed against CCP1 [6,7] of human DAF, recognized the proteins of chimpanzee and gorilla. The anity of IA10 for Old World monkey DAF was very low. 3.2. RT-PCR analysis of primate RNA Fig. 2 shows RT-PCR products corresponding to DAF's coding region (prepared from PBMCs of patas monkey and from cell lines of chimpanzee, gorilla, hamadryas baboon and Rhesus macaque). A band similar in size to that expected for human DAF is apparent for each of the species. At least one additional band, 100±200 bp smaller than that of human DAF, was recovered for chimpanzee and patas monkey. 3.3. Analyses of primate DAF cDNAs/amino acid sequences The derived amino acid sequences of the primate DAF proteins aligned using the CLUSTAL W program (Thompson JD, Higgins DG, Gibson

Fig. 1. Western blot analyses of primate erythrocyte DAF proteins. Panel A: Studies using rabbit polyclonal anti-Ehu DAF antibody and 7.5% SDS-PAGE. In addition to a DAF band between 60 and 70 kDa which is seen for all ®ve nonhuman primates, chimpanzee and gorilla exhibit the high molecular weight (dimeric) DAF form termed DAF2. Two bands considerably smaller than 60±70 kDa which are probably alternative splice products are seen for chimpanzee. Panel B: Studies using mouse IA10 mAb (against human CCP1) and 10% SDS-PAGE. Only one small chimpanzee DAF protein band is observed. The additional band at 100 kDa in the chimpanzee is faint in the original gel with the mAb and poorly observed on the scan.

Fig. 2. RT-PCR products of total RNA (prepared from PBMCs of patas monkey B289 and from cell lines of chimpanzee LENA, gorilla ROK, hamadryas baboon 26CB1 and Rhesus macaque NCTC Clone 3526). 5 ' and 3' primers are, respectively, D185 (B, C, M) (beginning of CCP1) and D1232R (in the 3' UTR). Chimpanzee and patas monkey each have three bands.

820

L. Kuttner-Kondo et al. / Developmental and Comparative Immunology 24 (2000) 815±827

Fig. 3. Derived amino acid sequences of G. gorilla (gorilla, GO), P. troglodytes (common chimpanzee, CH), P. hamadryas (hamadryas baboon, BA), M. mulatta (Rhesus macaque, RH), and E. patas (patas monkey, PA) DAF compared with those of human (HU), P. pygmaeus (orangutan, OR) and the GPI-anchoring forms of C. porcellus (guinea pig, GU), M. musculus (mouse, MO), and R. norvegius (rat, RA) DAF. Accession numbers for these sequences are given in Table 3. Alignment was made using CLUSTAL W (Thompson JD, Higgins DG, Gibson TJ, modi®ed, in Biology Workbench 3.2) [20±23] with modi®cation in the S/T domain. Numbering of the derived amino acid sequences begins with the ®rst amino acid of human CCP1 which corresponds to the start of the mature protein. The ®rst seven amino acids of CCP1 for orangutan were not determined [8]. Amino acid #338 in patas monkey is A or G. Key: () conserved residue in all species shown; (:) conservative replacements in all species shown; (+) conserved residues in primate DAF protein; N residue of potential N-glycosylation sites in bold; the predicted site of GPI-anchor attachment in human underlined. Primates are separated from other species by a line (±).

TJ, modi®ed, in Biology Workbench 3.2) [20±23] with subsequent manual adjustments are shown along with the corresponding GPI-anchored sequences of the guinea pig, mouse and rat in Fig. 3. The cDNA sequences can be found in the GenBank. Accession numbers of the major sequences are listed in Table 3 and other variants in Section 5. Gorilla, hamadryas baboon, and Rhesus macaque have one DAF mRNA similar to that of human DAF which codes for four

CCPs, the three S/T A, B and C regions, and the C-terminal hydrophobic signal sequence that directs GPI anchor addition. These and other domains in the DAF proteins from other species are summarized in Table 2. Chimpanzee and patas monkey have a similar full-length mRNA, but, in addition, have at least one smaller mRNA that does not include one or another portion of the S/T region (Fig. 4). The chimpanzee's short mRNA deletes S/T-A, while patas' short

L. Kuttner-Kondo et al. / Developmental and Comparative Immunology 24 (2000) 815±827

821

Fig. 3 (continued)

mRNA deletes S/T-A, S/T-B, and sometimes S/ T-C. Additionally, hamadryas baboon, Rhesus macaque and patas monkey mRNAs do not

encode a 12 base pair repeat (which translates to KTTT) in the S/T-A region, as well as four bases in the 3' UTR (see GenBank). Of the two patas

822

L. Kuttner-Kondo et al. / Developmental and Comparative Immunology 24 (2000) 815±827

Table 2 Conservation of DAF domainsa Domain

CCPs ST GPI Transmembrane Secretory Potential N-glycosylation sites CCP1 CCP1-2 linker CCP2 CCP3 CCP4 a

Species HU

CH

GO

OR

RH

BA

PA

GU

MO

RA

+ + +

+ V +

+ + +

+ + +

+ + +

+ + +

+ + +

+ V + + +

+ + + +

+ V + + +

+

+

+ +

+

+ +

+

? +

+ + +

+ + + +

+

+: Present; V: variable; ?: unknown.

monkeys sequenced, one has a G at amino acid 338, while the other has an A, indicating a polymorphism at this position. The percent nucleotide identity to the coding region of human DAF for the full-length form of

chimpanzee is 99%, gorilla 99%, hamadryas baboon 91%, Rhesus macaque 92%, and patas monkey 91%. The corresponding percent amino acid identities to human DAF protein of the derived amino acid sequences are chimpanzee 98%,

Fig. 4. Schematic diagram of the mature DAF protein in non-human primates and alternative splice products in chimpanzee and patas monkey.

L. Kuttner-Kondo et al. / Developmental and Comparative Immunology 24 (2000) 815±827

gorilla 97%, hamadryas baboon 82%, Rhesus macaque 84%, and patas monkey 83%. The derived amino acid sequences of the three Cercopithecine monkeys have higher homology to each other than to the great apes. In the CCP functional domains, the percent amino acid identities of the non-human primates, guinea pig, mouse and rat relative to human DAF are given in Table 3. The phylogenetic relationship of these primate and non-primate DAFs is shown in a PHYLIP unrooted tree [20±25] (Fig. 5). There are stretches of non-human primate amino acid sequence where R1 di€erence from human sequence is present (Fig. 3). These include the end of CCP1 and connecting linker, the linker between CCPs 2 and 3, the middle of CCP3, the end of CCP3 and beginning of CCP4, and the middle to end of CCP4. While 040% of the derived amino acids outside these regions would be considered conservative replacements, the remaining 60% are not. In particular, in CCPs 2 and 3, four P residues in human DAF are L, A, S or T in DAF from the Old World monkeys. Interestingly, three non-P residues (T, R, and S) in human CCPs 2 and 3 are P in the Old World monkeys. These amino acid variations may relate to structural variations that have evolved in concert with evolution of the C3 convertases in these primates. The N-linked glycosylation site in human DAF at amino acid position 61 between CCPs 1 and 2

823

is conserved in all ®ve non-human primates. The cloned Rhesus DAF protein when expressed in a COS cell system appeared to primarily produce a glycosylated form (93% by densitometry; Kuttner-Kondo and Medof, unpublished data). In hamadryas baboon, another potential N-linked glycosylation site is found at position 34 in CCP1 (N34 to T36). F is the intermediate residue at position 35. Based on the molecular weight (Fig. 1), it is unlikely that both sites in baboon are utilized at the same time. 3.4. 5 ' signal sequences of primate DAF In contrast to the 34-amino acid long signal sequence in human DAF cDNA, patas monkey has a signal sequence of 33 amino acids, baboon has 34 amino acids, and Rhesus macaque has 35 amino acids. Preliminary sequencing showed few amino acid di€erences between the non-human primate and human leader sequences. The data indicated that (1) Rhesus macaque and baboon substitute an L for V at (human) leader position #25, while patas monkey drops the codon; (2) all three Old World monkeys substitute an A for G at #34, (3) Rhesus macaque adds an L following leader position #25, and (4) patas monkey B289 substitutes a P for A at leader position #4, an L for P at #9, and a P for A at #10.

Table 3 Comparison of DAF amino acids in CCPs 1±4a Species

Accession No.

Percent identity

Human Chimpanzee Gorilla Orangutan Rhesus monkey Hamadryas baboon Patas monkey (B289) Guinea pig Mouse, gpi Rat

M31516 AF149760 AF149759 S67775 AF149763 AF149762 AF149764 D49416 L41366 AF039583

100.0 98.4 97.6 93.0 83.6 82.4 83.2 58.5 53.8 52.4

a

Based on Fig. 3 alignment.

Fig. 5. PHYLIP unrooted tree based on the alignments shown in Fig. 3 beginning with human residue #8 [20±25].

824

L. Kuttner-Kondo et al. / Developmental and Comparative Immunology 24 (2000) 815±827

3.5. 44 base pair deletion in CCP3 of the Rhesus macaque OPE340 In the course of sequencing RT-PCR products from PBMCs of Rhesus macaque, a sequence was found paralleling the alternative splicing of human DAF mRNA that occurs in the Dr(a-) Cromer blood group phenotype [26]. In humans, a C to T transversion at base 678 (from transcriptional start site) [27], which results in L in place of S at position 165, is associated with a 44-base pair deletion in CCP3 that results in a frameshift and early stop codon. Rhesus macaques have a P at position 165. 4. Discussion In this study, we cloned and characterized DAF cDNAs expressed by ®ve non-human primates Ð two great apes and three Old World monkeys. We found that while all expressed DAF proteins corresponding to 70 kDa human DAF, two expressed additional mRNAs varying in their S/T-rich regions. Interestingly in one case, the Rhesus macaque, a DAF variant paralleling the Dr(a-) Cromer phenotype was found. In all cases, only DAF sequences predictive of GPI-anchored proteins were identi®ed. 4.1. Homology The molecular cloning of the primate DAF cDNAs revealed high homology between human, chimpanzee and gorilla DAF proteins and lower homology between human/great ape DAF and Old World monkeys' DAF (Table 3, Fig. 3). This was expected based on estimated divergence dates [28] and is in accordance with sequence information on primate MCP [29]. The DAF phylogenetic tree (Fig. 5) supports the catarrhine (hominoids and Old World monkey) primate evolution hypothesis [28]. Based on the high homology between human and orangutan DAF, it was expected that chimpanzee and gorilla, who are regarded as diverging later than orangutan from humans, would also show high homology to human DAF. Slightly lower homology would

be expected for the Old World monkeys which diverged 025 million years ago from the hominoids [28]. 4.2. CCPs In humans, DAF's function has been assigned to CCPs 2 and 3 for the classical pathway, and to CCPs 2, 3 and 4 for the alternative pathway [7]. At the primary amino acid level, the stretches of highest (no di€erences) conservation include the end of CCP1 and attached linker, the linker region between CCPs 2 and 3, the middle of CCP3, the beginning of CCP4, and the late middle of CCP4 (Fig. 3). These areas may be critical for DAF's binding to the respective species' C3 and C5 convertases. Notable residues may be (1) the three consecutive lysines (in human, KKK125±127) between CCPs 2 and 3 which are preserved in all ®ve primates, (2) the LF147,148 residues in CCP3 which constitute part of a hydrophobic pocket, and (3) the middle of CCPs 3 and 4 which are highly conserved and have high homology to a ``repeat'' found in the vaccinia virus complement control protein which is reported to have classical pathway decay-accelerating activity on human complement [30±32]. Recent structure/function studies [33] have shown that mutation of KKK125±127 abolishes alternative regulatory pathway activity, while mutation of LF147,148 abolishes classical regulatory activity and has a marked negative e€ect on alternative pathway activity. Some of these residues may be important to other regulators of complement activation. Substitution of F82V in CCP2 of human CR1, analogous to DAF's F148, nearly abolished CR1's decay activity in both pathways [34]. 4.3. S/T region Human, gorilla, Rhesus macaque and hamadryas baboon express mRNA transcripts for only one DAF form that includes the complete S/T domain denoted as S/T-A, B, C in humans. In contrast, chimpanzee has mRNA encoding at least two forms: a long form which has S/T-A, B, C, and a short form which has S/T-B, C (Fig. 4).

L. Kuttner-Kondo et al. / Developmental and Comparative Immunology 24 (2000) 815±827

Therefore, the appearance of this apparently alternative splice variant must have occurred after human/chimpanzee divergence. However, the S/T-A nucleotide sequence in human and chimpanzee is the same. Consequently, an explanation of the molecular underpinnings of the alternative splice decision may lie within the preceding intronic sequence. Patas monkey has a full-length DAF form containing S/T-A, B, C. However, this monkey also has mRNAs for short forms that do not include S/T-A, B, or S/T-A, B, C. While a chimpanzee short form protein is seen on the Western blot of its erythrocyte lysate using polyclonal antibodies, as well as mAb IA10 to Ehu DAF (which indicates that the chimpanzee's short form is expressed), a patas short form is not seen on the same Western blot. In humans, the GPI anchor is thought to attach to the last residue of the S/ T-B region. In patas, it therefore is possible that: (1) only patas long form is expressed in erythrocytes, (2) an alternative o site, amino acid site of attachment of GPI-anchor donor, is chosen and a GPI anchor is attached to the short form in some cell types, (3) the short form protein is retained and degraded in the endoplasmic reticulum due to a non-functional GPI signal [35], or (4) the short form protein is secreted although the C-terminal sequence contains a free C residue that is associated with ER retention [35]. It may be seen in Fig. 3 that the S/T-A region, especially the portion close to CCP4, has more variability than other regions of DAF across the mammals sequenced. Consistent with this, guinea pig DAF with naturally occurring shorter S/T regions has been shown by 51Cr release and C3 deposition assays to be less ecient in inhibiting the classical and alternative pathway convertases than the naturally occurring longer forms [13]. The spacer provided by the S/T region also is functionally important in the binding of Dr and certain Dr adhesin1-related X adhesins of E. coli 1

E. coli bearing Dr and Dr-related X adhesins bind to DAF, leading to intestinal and urinary tract infections [36,37]. 2 The upstream (a) and downstream (b) positions of CCP3 are encoded by di€erent exons [17].

825

[36,37] and the binding of echovirus 7 [38]. The biologic consequences of the shorter DAF forms requires further investigation. 4.4. Dr(a-) phenotype (in human, S165L) In humans, the presence of a T in place of a C at nucleotide position 678 predisposes to abnormal splicing of 44 bases at the beginning of CCP3b2 (exon 5), generating a stop codon [26]. This resulting change of S165 to L is responsible for the absence of the Cromer blood group antigen Dr(a), termed the Dr(a-). The Old World monkeys in this study and the orangutan have a P at the equivalent position [8]. This di€erence alone would probably convert these primates to ``Dr(a-)''. The Rhesus macaque was found to exhibit a phenomenon similar to Dr(a-) in humans, in that a 44-nucleotide deletion was found in CCP3 which led to a frameshift and early stop codon. The nucleotide change that triggers the alternative splicing event in humans may be di€erent from that in the Rhesus macaque. It is not known if the same phenomenon of reduced DAF expression occurs in the monkey.

5. Accession numbers The following GenBank accession numbers have been assigned to the nucleotide sequences reported here: AF149759 (Gorilla gorilla ROK), AF149760-1 (Pan troglodytes LENA), AF149762 (Papio hamadryas 26CB1), AF149763 (Macaca mulatta NCTC clone 3526), AF149764-6 (Erythrocebus patas B289), and AF149767 (Erythrocebus patas R181).

Acknowledgements This investigation was supported from grants from the National Institutes of Health (NIH) R01 AI23598 and T32 CA73515. The authors thank Sara Cechner for manuscript preparation.

826

L. Kuttner-Kondo et al. / Developmental and Comparative Immunology 24 (2000) 815±827

References [1] Medof ME, Walter EI, Rutgers JL, Knowles DM, Nussenzweig V. Identi®cation of the complement decayaccelerating factor (DAF) on epithelium and glandular cells and in body ¯uids. J Exp Med 1987;165:848±64. [2] Fujita T, Inoue T, Ogawa K, Iida K, Tamura N. The mechanism of action of decay-accelerating factor (DAF). DAF inhibits the assembly of C3 convertases by dissociating C2a and Bb. J Exp Med 1987;166:1221±8. [3] Medof ME, Walter EI, Roberts WL, Haas R, Rosenberry TL. Decay-accelerating factor of complement is anchored to cells by a C-terminal glycolipid. Biochemistry 1986;25:6740±7. [4] Medof ME, Lublin DM, Holers VM, Ayers DJ, Getty RR, Leykam JF, Atkinson JP, Tykocinski ML. Cloning and characterization of cDNAs encoding the complete sequence of decay-accelerating factor of human complement. Proc Natl Acad Sci USA 1987;84:2007±11. [5] Lublin DM, Krsek-Staples J, Pangburn MK, Atkinson JP. Biosynthesis and glycosylation of the complement regulatory protein decay accelerating factor. J Immunol 1986;137:1629±35. [6] Coyne KE, Hall SE, Thompson ES, Arce MA, Kinoshita T, Fujita T, Anstee DJ, Rosse W, Lublin DM. Mapping of epitopes, glycosylation sites, and complement regulatory domains in human decay accelerating factor. J Immunol 1992;149:2906±13. [7] Brodbeck WG, Liu D, Sperry J, Mold C, Medof ME. Localization of classical and alternative pathway regulatory activity within the decay-accelerating factor. J Immunol 1996;156:2528±33. [8] Nickells MW, Alvarez JI, Lublin DM, Atkinson JP. Characterization of DAF-2, a high molecular weight form of decay-accelerating factor (DAF; CD55), as a covalently cross-linked dimer of DAF-1. J Immunol 1994;152:676±85. [9] Spicer AP, Seldin MF, Gendler SJ. Molecular cloning and chromosomal localization of the mouse decayaccelerating factor genes. Duplicated genes encode glycosylphosphatidylinositol-anchored and transmembrane forms. J Immunol 1995;155:3079±91. [10] Nonaka M, Miwa T, Okada N, Nonaka M, Okada H. Multiple isoforms of guinea pig decay-accelerating factor (DAF) generated by alternative splicing. J Immunol 1995;155:3037±48. [11] Hinchli€e SJ, Spiller OB, Rushmere NK, Morgan BP. Molecular cloning and functional characterization of the rat analogue of human decay-accelerating factor (CD55). J Immunol 1998;161:5695±703. [12] Sta€ord HA, Tykocinski ML, Lublin DM, Holers VM, Rosse WF, Atkinson JP, Medof ME. Normal polymorphic variations and transcription of the decay-accelerating factor (DAF) gene in normal and paroxysmal nocturnal hemoglobinuria (PNH) cells. Proc Natl Acad Sci USA 1988;85:880±4. [13] Wang G, Nonaka M, He C, Okada N, Nakashima I,

[14]

[15]

[16]

[17]

[18]

[19]

[20] [21]

[22] [23]

[24] [25] [26]

[27]

Okada H. Functional di€erences among multiple isoforms of guinea pig decay-accelerating factor. J Immunol 1998;160:3014±22. Miwa T, Okada N, Okada H. Alternative exon usage in the 3' region of a single gene generates glycosyl-phosphatidylinositol-anchored and transmembrane forms of rat decay-accelerating factor. Immunogenetics 2000;51:129± 37. Caras IW, Davitz MA, Rhee L, Weddell G, Martin Jr. DW. Cloning of decay accelerating factor suggests novel use of splicing to generate two proteins. Nature 1987;325:545±9. Nonaka M, Nonaka M, Takenaka O, Okada N, Okada H. A new repetitive sequence uniquely present in the decay-accelerating factor genes. Immunogenetics 1998;47:246±55. Post TW, Arce MA, Liszewski MK, Thompson ES, Atkinson JP, Lublin DM. Structure of the gene for human complement protein decay accelerating factor. J Immunol 1990;144:740±4. Kuttner-Kondo L, Medof ME, Brodbeck W, Shoham M. Molecular modeling and mechanism of action of human decay-accelerating factor. Prot Eng 1996;9:1143± 9. Kinoshita T, Medof ME, Silber R, Nussenszweig V. Distribution of decay-accelerating factor in the peripheral blood of normal individuals and patients with paroxysmal nocturnal hemoglobinuria. J Exp Med 1985;162:75± 92. Higgins DG, Bleasby AJ, Fuchs R. CLUSTAL V: improved software for multiple sequence alignment. CABIOS 1992;8:189±91. Thompson JD, Higgins DG, Gibson TJ. CLUSTAL W: improving the sensitivity of progressive multiple sequence alignment through sequence weighting, position-speci®c gap penalties and weight matrix choice. Nucleic Acids Res 1994;22:4673±80. Subramaniam S. Biology workbench Ð a seamless database and analysis environment for the biologist. Proteins 1998;32:1±2. Unwin R, Fenton J, Whitsitt M, Jamison C, Stupar M, Jakobsson E, Subramaniam S. Biology workbench: a computing and analysis environment for the biological sciences. In: Letovsky S, editor. Bioinformatics: databases and systems. Boston: Kluwer Academic Publishers, 1999. p. 233±44. Felsenstein J. PHYLIP Ð Phylogeny Inference Package (Version 3.2). Cladistics 1989;5:164±6. Felsenstein J. PHYLIP (Phylogeny Inference Package), Version 3.5c. Seattle: Department of Genetics, University of Washington, 1993. Lublin DM, Mallison G, Poole J, Reid ME, Thompson ES, Ferdman BR, Telen MJ, Anstee DJ, Tanner MJA. Molecular basis of reduced or absent expression of decay-accelerating factor in Cromer blood group. Blood 1994;84:1276±87. Ewulonu UK, Ravi L, Medof ME. Characterization of

L. Kuttner-Kondo et al. / Developmental and Comparative Immunology 24 (2000) 815±827

[28] [29]

[30] [31]

[32]

[33]

the decay-accelerating factor gene promoter region. Proc Natl Acad Sci USA 1991;88:4675±9. Stewart C-B, Disotell TR. Primate evolution Ð in and out of Africa. Curr Biol 1998;8:R582±R588. Hsu EC, DoÈrig RE, Sarangi F, Marcil A, Iorio C, Richardson CD. Arti®cial mutations and natural variations in the CD46 molecules from human and monkey cells de®ne regions important for measles virus binding. J Virol 1997;71:6144±54. Kotwal GJ, Moss B. Vaccinia virus encodes a secretory polypeptide structurally related to complement control proteins. Nature 1988;335:176±8. Kotwal GJ, Isaacs SN, McKenzie R, Frank MM, Moss B. Inhibition of the complement cascade by the major secretory protein of vaccinia virus. Science 1995;250:827± 30. Isaacs SN, Kotwal GJ, Moss B. Vaccinia virus complement-control protein prevents antibody-dependent complement-enhanced neutralization of infectivity and contributes to virulence. Proc Natl Acad Sci USA 1992;89:628±32. Brodbeck WG, Kuttner-Kondo L, Yu J, Medof ME. In: XVIIth International Complement Workshop, Rhodes, Greece. 1998.

827

[34] Krych-Goldberg M, Hauhart RE, Subramanian VB, Yurcisin II BM, Crimmins DL, Hourcade DE, Atkinson JP. Decay accelerating activity of complement receptor type 1 (CD35). J Biol Chem 1999;274:31160±8. [35] Field MC, Moran P, Li W, Keller G-A, Caras IW. Retention and degradation of proteins containing an uncleaved glycosylphosphatidylinositol signal. J Biol Chem 1994;269:10830±7. [36] Pham T, Kaul A, Hart A, Golusko P, Moulds J, Nowicki S, Lublin DM, Nowicki BJ. dra-related X adhesins of gestational pyelonephritis-associated Escherichia coli recognize SCR-3 and SCR-4 domains of recombinant decay -accelerating factor. Infect Immun 1995;63:1663±8. [37] Nowicki B, Hart A, Coyne KE, Lublin DM, Nowicki S. Short consensus repeat-3 domain of recombinant decayaccelerating factor is recognized by Escherichia coli recombinant Dr adhesin in a model of a cell-cell interaction. J Exp Med 1993;178:2115±21. [38] Clarkson NA, Kaufman R, Lublin DM, Ward T, Pipkin PA, Minor PD, Evans DJ, Almond JW. Characterization of the echovirus 7 receptor: domains of CD55 critical for virus binding. J Virol 1995;69:5497±501.